Multilevel diffusion tensor imaging classification technique for characterizing neurobehavioral disorders

  • Josué Luiz Dalboni da Rocha
  • Gabriel Coutinho
  • Ivanei Bramati
  • Fernanda Tovar Moll
  • Ranganatha SitaramEmail author


This proposed novel method consists of three levels of analyses of diffusion tensor imaging data: 1) voxel level analysis of fractional anisotropy of white matter tracks, 2) connection level analysis, based on fiber tracks between specific brain regions, and 3) network level analysis, based connections among multiple brain regions. Machine-learning techniques of (Fisher score) feature selection, (Support Vector Machine) pattern classification, and (Leave-one-out) cross-validation are performed, for recognition of the neural connectivity patterns for diagnostic purposes. For validation proposes, this multilevel approach achieved an average classification accuracy of 90% between Alzheimer’s disease and healthy controls, 83% between Alzheimer’s disease and mild cognitive impairment, and 83% between mild cognitive impairment and healthy controls. The results indicate that the multilevel diffusion tensor imaging approach used in this analysis is a potential diagnostic tool for clinical evaluations of brain disorders. The presented pipeline is now available as a tool for scientifically applications in a broad range of studies from both clinical and behavioral spectrum, which includes studies about autism, dyslexia, schizophrenia, dementia, motor body performance, among others.


Diffusion tensor imaging Fractional anisotropy Fiber tracking Graph theory Machine learning 



The senior author of this study was supported by the Indigo Project FKZ 01DQ13004, and Fondecyt Regular projects number 1171313 and number 1171320.

Compliance with ethical standards

Ethical approval

All procedures involving human participants were in accordance with the ethical standards of the institutional research committee and with the 1964 Helsinki declaration.

Conflict of interest

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.


  1. Ashburner, J., Barnes, G., Chen, C., et al. (2010). SPM 8 Manual. Accessed 09 September 2018.
  2. Barthelemy, M. (2004). Betweenness centrality in large complex networks. The European Physical Journal B, 38(2), 163–168.CrossRefGoogle Scholar
  3. Basser, P. J., & Pierpaoli, C. (1996). Microstructural and physiological features of tissues elucidated by quantitative-diffusion-tensor MRI. Journal of Magnetic Resonance. Series B, 111, 209–219.CrossRefPubMedGoogle Scholar
  4. Becker, B., & Drechsler, R. (1998). Binary decision diagrams: Theory and implementation. Springer. isbn:978-1-4419-5047-5.Google Scholar
  5. Bihan, D. L., & Breton, E. (1985). Imagerie de diffusion in-vivo par résonance. Comptes rendus de l'Académie des Sciences., 301(15), 1109–1112.Google Scholar
  6. Brodmann, K. (1909). Vergleichende Lokalisationslehre der Grosshirnrinde. Leipzig: Johann Ambrosius Barth.Google Scholar
  7. Cumming, G., & Calin-Jageman, R. (2017). Introduction to the new statistics. New York: Routledge.Google Scholar
  8. Demirhan, A., Nir, T. M., Zavaliangos-Petropulu, A. (2015). Feature selection improves the accuracy of classifying Alzheimer disease using diffusion tensor images. IEEE 12th International Symposium on Biomedical Imaging, New York.
  9. Dyrba, M., Ewers, M., Wegrzyn, M., Kilimann, I., Plant, C., Oswald, A., Meindl, T., Pievani, M., Bokde, A. L. W., Fellgiebel, A., Filippi, M., Hampel, H., Klöppel, S., Hauenstein, K., Kirste, T., Teipel, S. J., & the EDSD study group. (2013). Robust automated detection of classification of multicenter DTI data. PLoS One, 8(5), e64925. Scholar
  10. Ebadi, A., Dalboni da Rocha, J. L., Nagaraju, D. B., Tovar-Moll, F., Bramati, I., Coutinho, G., Sitaram, R., & Rashidi, P. (2017). Ensemble classification of Alzheimer's disease and mild cognitive impairment based on complex graph measures from diffusion tensor images. Frontiers in Neuroscience, 11, 56. Scholar
  11. Einstein, A. (1956). Investigations on the theory of Brownian motion. New York: Dover.Google Scholar
  12. Fagiolo, G. (2007). Clustering in complex directed networks. Physical Review E, 76(2), 026107.CrossRefGoogle Scholar
  13. Foster, E. D., & Deardorff, A. (2017). Open science framework (OSF). Journal of the Medical Library Association, 105(2), 203–206. Scholar
  14. Friston, K. J. (1996). Statistical parametric mapping and other analysis of functional imaging data. Brain Mapping: The Methods, pages 363–385. Academic Press.Google Scholar
  15. Gong, G., He, Y., Concha, L., Lebel, C., Gross, D., Evans, A., & Beaulieu, C. (2008). Mapping anatomical connectivity patterns of human cerebral cortex using in vivo diffusion tensor imaging tractography. Cerebral Cortex, 19(3, 1 March 2009), 524–536. Scholar
  16. Good, P. I. (2000). Permutation tests: A practical guide to resampling methods for testing hypotheses, springer series in statistics (Vol. 2). Springer.Google Scholar
  17. He, X., Cai, D., & Niyogi, P. (2005). Laplacian score for feature selection. Advances in Neural Information Processing Systems, (pp. 507–514).Google Scholar
  18. Hong, S., Bernhardt, B. C., Caldairou, B., Hall, J. A., Guiot, M. C., Schrader, D., Bernasconi, N., & Bernasconi, A. (2017). Multimodal MRI profiling of focal cortical dysplasia type II. Neurology, 88(8), 734–742. Scholar
  19. Huettel, S. A., Song, A. W., & McCarthy, G. (2004). Functional magnetic resonance imaging. Sunderland, Massachusetts: Sinauer Associates Publishers.Google Scholar
  20. Jin, B., Strasburger, A., Laken, S. J., Kozel, F. A., Johnson, K. A., et al. (2009). Feature construction and selection for fMRI-based deception detection. BMC Bioinformatics, 10(Suppl 9), S15.CrossRefPubMedPubMedCentralGoogle Scholar
  21. Johnson, R., & Freund, J. (2011). Miller and Freund’s probability and statistics for engineers (8th ed.) Prentice Hall International.Google Scholar
  22. Latora, V., & Marchiori, M. (2001). Efficient behavior of small-world networks. Physics Review, 87, 198701.Google Scholar
  23. Lawes, I. N., Barrick, T. R., Murugam, V., Spierings, N., Evans, D. R., et al. (2008). Atlas-based segmentation of white matter tracts of the human brain using diffusion tensor tractography and comparison with classical dissection. Neuroimage, 39, 62–79.CrossRefPubMedGoogle Scholar
  24. Li, M., Qin, Y., Gao, F., Zhu, W., & He, X. (2014a). Discriminative analysis of multivariate features from structural MRI and diffusion tensor images. Magnetic Resonance Imaging, 32, 1043–1051. Scholar
  25. Li, Q., Rajagopalan, C., & Clifford, G. D. (2014b). A machine learning approach to multi-level ECG signal quality classification. Computer Methods and Programs in Biomedicine, 117(3), 435–447. Scholar
  26. Ojola, M., & Garriga, G. C. (2010). Permutation tests for studying classifier performance. Journal of Machine Learning Research, 11, 1833–1863.Google Scholar
  27. Pierpaoli, C., & Basser, P. J. (1996). Toward a quantitative assessment of diffusion anisotropy. Magnetic Resonance in Medicine, 36(6), 893–906.CrossRefPubMedGoogle Scholar
  28. Radmacher, M. D., McShane, L. M., & Simon, R. (2002). A paradigm for class prediction using gene expression profiles. Journal of Computational Biology, 9, 505–512.CrossRefPubMedGoogle Scholar
  29. Rosen, K. H. (2006). Discrete mathematics and its applications. McGraw-Hill.Google Scholar
  30. Sporns, O. (2003). Graph theory methods for the analysis of neural connectivity patterns. Neuroscience Databases, 171–186.Google Scholar
  31. Vapnik, V., & Lerner, A. (1963). Pattern recognition using generalized portrait method. Automation and Remote Control, 24, 774–780.Google Scholar
  32. Varkuti, B., Cavusoglu, M., Kullik, A., Schiffler, B., Veit, R., Yilmaz, O., et al. (2011). Quantifying the link between anatomical connectivity, gray matter volume and regional cerebral blood flow: An integrative MRI study. PLoS One, 6(4), e14801.CrossRefPubMedPubMedCentralGoogle Scholar
  33. Yeh, F. C., & Tseng, W. Y. (2011). NTU-90: A high angular resolution brain atlas constructed by q-space diffeomorphic reconstruction. Neuroimage, 58, 91–99.CrossRefGoogle Scholar
  34. Yeh, F. C., Verstynen, T. D., Wang, Y., Fernández-Miranda, J. C., & Tseng, W. I. (2013). Deterministic diffusion fiber tracking improved by quantitative anisotropy. PLoS One, 8(11), e80713. Scholar
  35. Zurita, M., Montalba, C., Labbé, T., Cruz, J. P., Dalboni da Rocha, J., Tejos, C., Ciampi, E., Cárcamo, C., Sitaram, R., et al. (2018). Characterization of relapsing-remitting multiple sclerosis patients using support vector machine classifications of functional and diffusion MRI data. NeuroImage Clinical, 20, 724–730. Advance online publication. Scholar

Copyright information

© Springer Science+Business Media, LLC, part of Springer Nature 2018

Authors and Affiliations

  1. 1.Brain and Language Lab, Department of Clinical NeuroscienceUniversity of GenevaGenevaSwitzerland
  2. 2.Department of Biomedical EngineeringUniversity of FloridaGainesvilleUSA
  3. 3.D’Or Institute for Research and EducationRio de JaneiroBrazil
  4. 4.Federal Univerisity of Rio de JaneiroRio de JaneiroBrazil
  5. 5.Institute for Biological and Medical Engineering, Schools of Engineering, Biology and MedicinePontificia Universidad Católica de ChileSantiagoChile
  6. 6.Department of Psychiatry and Section of Neuroscience, School of MedicinePontificia Universidad Católica de ChileSantiagoChile
  7. 7.Laboratory for Brain-Machine Interfaces and NeuromodulationPontificia Universidad Católica de ChileSantiagoChile

Personalised recommendations